<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ramesh-etal-2024-gpt">
    <titleInfo>
        <title>GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Govind</namePart>
        <namePart type="family">Ramesh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yao</namePart>
        <namePart type="family">Dou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wei</namePart>
        <namePart type="family">Xu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yaser</namePart>
            <namePart type="family">Al-Onaizan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohit</namePart>
            <namePart type="family">Bansal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yun-Nung</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Research on jailbreaking has been valuable for testing and understanding the safety and security issues of large language models (LLMs). In this paper, we introduce Iterative Refinement Induced Self-Jailbreak (IRIS), a novel approach that leverages the reflective capabilities of LLMs for jailbreaking with only black-box access. Unlike previous methods, IRIS simplifies the jailbreaking process by using a single model as both the attacker and target. This method first iteratively refines adversarial prompts through self-explanation, which is crucial for ensuring that even well-aligned LLMs obey adversarial instructions. IRIS then rates and enhances the output given the refined prompt to increase its harmfulness. We find that IRIS achieves jailbreak success rates of 98% on GPT-4, 92% on GPT-4 Turbo, and 94% on Llama-3.1-70B in under 7 queries. It significantly outperforms prior approaches in automatic, black-box, and interpretable jailbreaking, while requiring substantially fewer queries, thereby establishing a new standard for interpretable jailbreaking methods.</abstract>
    <identifier type="citekey">ramesh-etal-2024-gpt</identifier>
    <location>
        <url>https://aclanthology.org/2024.emnlp-main.1235</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>22139</start>
            <end>22148</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation
%A Ramesh, Govind
%A Dou, Yao
%A Xu, Wei
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F ramesh-etal-2024-gpt
%X Research on jailbreaking has been valuable for testing and understanding the safety and security issues of large language models (LLMs). In this paper, we introduce Iterative Refinement Induced Self-Jailbreak (IRIS), a novel approach that leverages the reflective capabilities of LLMs for jailbreaking with only black-box access. Unlike previous methods, IRIS simplifies the jailbreaking process by using a single model as both the attacker and target. This method first iteratively refines adversarial prompts through self-explanation, which is crucial for ensuring that even well-aligned LLMs obey adversarial instructions. IRIS then rates and enhances the output given the refined prompt to increase its harmfulness. We find that IRIS achieves jailbreak success rates of 98% on GPT-4, 92% on GPT-4 Turbo, and 94% on Llama-3.1-70B in under 7 queries. It significantly outperforms prior approaches in automatic, black-box, and interpretable jailbreaking, while requiring substantially fewer queries, thereby establishing a new standard for interpretable jailbreaking methods.
%U https://aclanthology.org/2024.emnlp-main.1235
%P 22139-22148

Markdown (Informal)

[GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation](https://aclanthology.org/2024.emnlp-main.1235) (Ramesh et al., EMNLP 2024)

• GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation (Ramesh et al., EMNLP 2024)

ACL

• Govind Ramesh, Yao Dou, and Wei Xu. 2024. GPT-4 Jailbreaks Itself with Near-Perfect Success Using Self-Explanation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 22139–22148, Miami, Florida, USA. Association for Computational Linguistics.